Radulescu, Eugenia, Ganeshan, Balaji, Shergill, Sukhwinder S, Medford, Nick, Chatwin, Chris, Young, Rupert and Critchley, Hugo (2014) Grey-matter texture abnormalities and reduced hippocampal volume are distinguishing features of schizophrenia. Psychiatry Research: Neuroimaging, 223 (3). pp. 179-186. ISSN 0925-4927

Preview

PDF (Authors accepted manuscript) - Accepted Version Download (692kB) | Preview

Abstract

Neurodevelopmental processes are widely believed to underlie schizophrenia. Analysis of
brain texture from conventional magnetic resonance imaging (MRI) can detect disturbance in
brain cytoarchitecture. We tested the hypothesis that patients with schizophrenia manifest
quantitative differences in brain texture that, alongside discrete volumetric changes, may
serve as an endophenotypic biomarker. Texture analysis (TA) of grey matter distribution and
voxel-based morphometry (VBM) of regional brain volumes were applied to MRI scans of 27
patients with schizophrenia and 24 controls. Texture parameters (uniformity and entropy)
were also used as covariates in VBM analyses to test for correspondence with regional brain
volume. Linear discriminant analysis tested if texture and volumetric data predicted
diagnostic group membership (schizophrenia or control). We found that uniformity and
entropy of grey matter differed significantly between individuals with schizophrenia and
controls at the fine spatial scale (filter width below 2 mm). Within the schizophrenia group,
these texture parameters correlated with volumes of the left hippocampus, right amygdala
and cerebellum. The best predictor of diagnostic group membership was the combination of
fine texture heterogeneity and left hippocampal size. This study highlights the presence of
distributed grey-matter abnormalities in schizophrenia, and their relation to focal structural
abnormality of the hippocampus. The conjunction of these features has potential as a
neuroimaging endophenotype of schizophrenia.

Item Type:	Article
Keywords:	Endophenotype, Magnetic resonance imaging, Neurodevelopment, Psychosis
Schools and Departments:	School of Engineering and Informatics > Engineering and Design Brighton and Sussex Medical School > Neuroscience
Subjects:	R Medicine R Medicine > R Medicine (General) > R856 Biomedical engineering. Electronics. Instrumentation R Medicine > R Medicine (General) > R858 Computer applications to medicine. Medical informatics R Medicine > R Medicine (General) > R895 Medical physics. Medical radiology. Nuclear medicine T Technology
Depositing User:	Chris Chatwin
Date Deposited:	19 Jun 2015 12:05
Last Modified:	26 Jul 2017 00:11
URI:	http://srodev.sussex.ac.uk/id/eprint/50754

View download statistics for this item

📧 Request an update